Submitted by Donald French.
Cushing Hamlen from Medtronic, Inc. gave the Industrial Problem Seminar on October 17, 1997 at the IMA. This article is a summary of the talk and presents Dr. Hamlen's work at Medtronic as well as his views on mathematics in industry. In the seminar he provided an overview of the products made by Medtronic and its mission. He then presented an in-depth description of the Materials and Modeling group of which he is a member and some of the activities and problems they have been involved with. The major agenda of his seminar was to reveal the purpose, prerequisite skills, and duties of a mathematically inclined scientist in a company like Medtronic. He emphasized that such a scientist needs a strong education in all areas of physics.
Medtronic makes many different medical devices, has revenues in excess of two billion dollars, has 11,500 employees, and is an international company with nearly 45% of its business outside of the United States. They build and design implantable pacemakers and defibrillators. Both of these devices have a small programmable control unit with a special cable that connects into the heart. They are designed to work reliably for at least ten years. A pacemaker regulates the heart rate of a patient. A defibrillator is used on a patient whose heart has a tendency to beat in highly irregular patterns. This device gives a pulse to the heart to jar it into a regular rhythm. The computer logic in a defibrillator is necessarily very complex so it can give a proper interpretation of how a patientŐs heart is behaving.
Medtronic also builds devices for vascular surgery, cardiology, and neurology. In cardiology, pumps and oxygenators are built. These pumps act as a bypass during heart surgery. It is important for these devices to handle the blood gently minimizing high shear stresses which could degrade it. In neurology special implantable devices are made that inject certain drugs which can dramatically ease the tremors in Parkinsons' patients and those with related diseases.
Other products made by Medtronic include artificial heart valves and other implantable devices which inject drugs in the spinal cord to control back pain. They are also working to create aids for snoring, hiccups and other less serious discomforts.
Most of the devices made at Medtronic must be placed in the body in an environment which is very hostile. These devices must work reliably for ten years --- they can't break or wear. To produce such high quality items their behavior in the body must be well understood --- a task that requires mathematical modeling.
The Center for Biomaterials Research where Dr. Hamlen is employed is set up to provide corporate wide research and development leadership, expertise in Polymer technologies, surface modification, and materials and modeling. The duties of this group include basic research of processes as well as concept and prototype analysis with the goal of reducing the number of prototypes needed to build a high quality device. They also provide "after the fact" device analysis. Dr. Hamlen is primarily a consultant for other scientists in the company. He is called in to model and simulate the performance of devices in various stages of production or when specific problems arise that are not well understood.
The applications this group encounters cover a wide range of physics supporting Dr. Hamlen's theme that an industry scientist needs to have a strong background in all areas of physics. The areas they are involved with include
A stated purpose for Medtronic activities is to make a fair profit. This goal has a strong influence in the day-to-day work of the industry scientist. Since it is important to bring the products to market as quickly as possible to be competitive, Dr. Hamlen pointed out that time was the most important factor in guiding his work and not efficiency. Thus his job is to obtain suitable answers in a timely manner. Because of this, "person time" is considerably more important that computer time. A brute force solution that may expend many CRAY hours but was simple for Dr. Hamlen to code is far preferable to a solution involving a complex (perhaps efficient) code that requires alot of human intervention. He typically uses general purpose codes instead of writing his own and tends to use far finer meshes than would be required in order to provide insurance that the solutions are correct while limiting the number of computer experiments.
A summary of the important skills that an industry scientist working in mathematical modeling needs are
2 Example Problems
Dr. Hamlen was a consultant on an analysis of the flow of blood in a pump (see figure 1) used during heart surgery.
Earlier work had led to the design of a narrower flow path and eliminated a large recirculation zone (see figure 2). A more recent question involved the design of the cutwater which had been identified as the location of the highest shear component (see figure 3). The key question was: does the shape of the cutwater-strut gap affect shear. It was important to minimize shear since this can damage the blood being pumped.
Two possible solutions were considered, hand built and molded designs (see figure 4). Dr. Hamlen performed two fluid dynamics computations to provide helpful information. In both numerical experiments he examined fluid moving through a rectangular box with a protrusion inserted in the path of the fluid. Figure 5 has the x-y cross section of the box. In the first computation the vertical profile of the tongue was in the hand built shape and in the second it was in the molded shape. The computation showed that the shear was higher in the hand built case.
Hamlen's work stimulated further investigation. An experiment with prototypes and real blood showed that there was a 23% hemolysis reduction from using the molded design instead of the hand built. Future designs (of all types) will use this modification and a patent has been filed.
Dr. Hamlen pointed out that the cutwater project came at a very busy time. He had only one afternoon to work on it. This situation dramatically exemplifies the importance of time in the real world.
Another project Dr. Hamlen worked on involved a cardioplegia heat exchanger (see figure 6). The question was whether adjustments in dimensions of the base of the exchanger could provide a more even blood flow through the manifold. Numerical computations were done on this problem that involved porous media (the heat exchanger) and were in two dimensions. The results of several numerical experiments are shown in figure 7. Design 5 was selected based on these calculations. When an apparatus based on design 5 was built and experiments were run on it, the scientist working with the blood flow observed that the flow was actually more even that predicted by the numerical calculations. This work exemplifies the importance of simulation.
Dr. Hamlen received his Ph.D. in Chemical Engineering from the University of Minnesota. He joined the Center for BioMaterials Research at Medtronic Inc in 1994. Dr. Hamlen can be reached through e-mail: firstname.lastname@example.org